ModelArray: An R package for statistical analysis of fixel-wise data.

Diffusion MRI is the dominant non-invasive imaging method used to characterize white matter organization in health and disease. Increasingly, fiber-specific properties within a voxel are analyzed using fixels. While tools for conducting statistical analyses of fixel-wise data exist, currently available tools support only a limited number of statistical models. Here we introduce ModelArray, an R package for mass-univariate statistical analysis of fixel-wise data. At present, ModelArray supports linear models as well as generalized additive models (GAMs), which are particularly useful for studying nonlinear effects in lifespan data. In addition, ModelArray also aims for scalable analysis. With only several lines of code, even large fixel-wise datasets can be analyzed using a standard personal computer. Detailed memory profiling revealed that ModelArray required only limited memory even for large datasets. As an example, we applied ModelArray to fixel-wise data derived from diffusion images acquired as part of the Philadelphia Neurodevelopmental Cohort (n = 938). ModelArray revealed anticipated nonlinear developmental effects in white matter. Moving forward, ModelArray is supported by an open-source software development model that can incorporate additional statistical models and other imaging data types. Taken together, ModelArray provides a flexible and efficient platform for statistical analysis of fixel-wise data.

Voxel-wise intermodal coupling analysis of two or more modalities using local covariance decomposition.

When individual subjects are imaged with multiple modalities, biological information is present not only within each modality, but also between modalities - that is, in how modalities covary at the voxel level. Previous studies have shown that local covariance structures between modalities, or intermodal coupling (IMCo), can be summarized for two modalities, and that two-modality IMCo reveals otherwise undiscovered patterns in neurodevelopment and certain diseases. However, previous IMCo methods are based on the slopes of local weighted linear regression lines, which are inherently asymmetric and limited to the two-modality setting. Here, we present a generalization of IMCo estimation which uses local covariance decompositions to define a symmetric, voxel-wise coupling coefficient that is valid for two or more modalities. We use this method to study coupling between cerebral blood flow, amplitude of low frequency fluctuations, and local connectivity in 803 subjects ages 8 through 22. We demonstrate that coupling is spatially heterogeneous, varies with respect to age and sex in neurodevelopment, and reveals patterns that are not present in individual modalities. As availability of multi-modal data continues to increase, principal-component-based IMCo (pIMCo) offers a powerful approach for summarizing relationships between multiple aspects of brain structure and function. An R package for estimating pIMCo is available at: https://github.com/hufengling/pIMCo.

Developmental coupling of cerebral blood flow and fMRI fluctuations in youth.

The functions of the human brain are metabolically expensive and reliant on coupling between cerebral blood flow (CBF) and neural activity, yet how this coupling evolves over development remains unexplored. Here, we examine the relationship between CBF, measured by arterial spin labeling, and the amplitude of low-frequency fluctuations (ALFF) from resting-state magnetic resonance imaging across a sample of 831 children (478 females, aged 8-22 years) from the Philadelphia Neurodevelopmental Cohort. We first use locally weighted regressions on the cortical surface to quantify CBF-ALFF coupling. We relate coupling to age, sex, and executive functioning with generalized additive models and assess network enrichment via spin testing. We demonstrate regionally specific changes in coupling over age and show that variations in coupling are related to biological sex and executive function. Our results highlight the importance of CBF-ALFF coupling throughout development; we discuss its potential as a future target for the study of neuropsychiatric diseases.

Fully automated detection of paramagnetic rims in multiple sclerosis lesions on 3T susceptibility-based MR imaging.

BACKGROUND AND PURPOSE: The presence of a paramagnetic rim around a white matter lesion has recently been shown to be a hallmark of a particular pathological type of multiple sclerosis lesion. Increased prevalence of these paramagnetic rim lesions is associated with a more severe disease course in MS, but manual identification is time-consuming. We present APRL, a method to automatically detect paramagnetic rim lesions on 3T T2*-phase images. METHODS: T1-weighted, T2-FLAIR, and T2*-phase MRI of the brain were collected at 3T for 20 subjects with MS. The images were then processed with automated lesion segmentation, lesion center detection, lesion labelling, and lesion-level radiomic feature extraction. A total of 951 lesions were identified, 113 (12%) of which contained a paramagnetic rim. We divided our data into a training set (16 patients, 753 lesions) and a testing set (4 patients, 198 lesions), fit a random forest classification model on the training set, and assessed our ability to classify paramagnetic rim lesions on the test set. RESULTS: The number of paramagnetic rim lesions per subject identified via our automated lesion labelling method was highly correlated with the gold standard count per subject, r = 0.86 (95% CI [0.68, 0.94]). The classification algorithm using radiomic features classified lesions with an area under the curve of 0.82 (95% CI [0.74, 0.92]). CONCLUSION: This study develops a fully automated technique, APRL, for the detection of paramagnetic rim lesions using standard T1 and FLAIR sequences and a T2*phase sequence obtained on 3T MR images.

TAPAS: A Thresholding Approach for Probability Map Automatic Segmentation in Multiple Sclerosis.

Total brain white matter lesion (WML) volume is the most widely established magnetic resonance imaging (MRI) outcome measure in studies of multiple sclerosis (MS). To estimate WML volume, there are a number of automatic segmentation methods available, yet manual delineation remains the gold standard approach. Automatic approaches often yield a probability map to which a threshold is applied to create lesion segmentation masks. Unfortunately, few approaches systematically determine the threshold employed; many methods use a manually selected threshold, thus introducing human error and bias into the automated procedure. In this study, we propose and validate an automatic thresholding algorithm, Thresholding Approach for Probability Map Automatic Segmentation in Multiple Sclerosis (TAPAS), to obtain subject-specific threshold estimates for probability map automatic segmentation of T2-weighted (T2) hyperintense WMLs. Using multimodal MRI, the proposed method applies an automatic segmentation algorithm to obtain probability maps. We obtain the true subject-specific threshold that maximizes the Sørensen-Dice similarity coefficient (DSC). Then the subject-specific thresholds are modeled on a naive estimate of volume using a generalized additive model. Applying this model, we predict a subject-specific threshold in data not used for training. We ran a Monte Carlo-resampled split-sample cross-validation (100 validation sets) using two data sets: the first obtained from the Johns Hopkins Hospital (JHH) on a Philips 3 Tesla (3T) scanner (n = 94) and a second collected at the Brigham and Women's Hospital (BWH) using a Siemens 3T scanner (n = 40). By means of the proposed automated technique, in the JHH data we found an average reduction in subject-level absolute error of 0.1 mL per one mL increase in manual volume. Using Bland-Altman analysis, we found that volumetric bias associated with group-level thresholding was mitigated when applying TAPAS. The BWH data showed similar absolute error estimates using group-level thresholding or TAPAS likely since Bland-Altman analyses indicated no systematic biases associated with group or TAPAS volume estimates. The current study presents the first validated fully automated method for subject-specific threshold prediction to segment brain lesions.

Multiple Sclerosis Lesion Segmentation with Tiramisu and 2.5D Stacked Slices.

In this paper, we present a fully convolutional densely connected network (Tiramisu) for multiple sclerosis (MS) lesion segmentation. Different from existing methods, we use stacked slices from all three anatomical planes to achieve a 2.5D method. Individual slices from a given orientation provide global context along the plane and the stack of adjacent slices adds local context. By taking stacked data from three orientations, the network has access to more samples for training and can make more accurate segmentation by combining information of different forms. The conducted experiments demonstrated the competitive performance of our method. For an ablation study, we simulated lesions on healthy controls to generate images with ground truth lesion masks. This experiment confirmed that the use of 2.5D patches, stacked data and the Tiramisu model improve the MS lesion segmentation performance. In addition, we evaluated our approach on the Longitudinal MS Lesion Segmentation Challenge. The overall score of 93.1 places the L 2-loss variant of our method in the first position on the leaderboard, while the focal-loss variant has obtained the best Dice coefficient and lesion-wise true positive rate with 69.3% and 60.2%, respectively.

A dual modeling approach to automatic segmentation of cerebral T2 hyperintensities and T1 black holes in multiple sclerosis.

BACKGROUND AND PURPOSE: Magnetic resonance imaging (MRI) is crucial for in vivo detection and characterization of white matter lesions (WML) in multiple sclerosis (MS). The most widely established MRI outcome measure is the volume of hyperintense lesions on T2-weighted images (T2L). Unfortunately, T2L are non-specific for the level of tissue destruction and show a weak relationship to clinical status. Interest in lesions that appear hypointense on T1-weighted images (T1L) ("black holes") has grown because T1L provide more specificity for axonal loss and a closer link to neurologic disability. The technical difficulty of T1L segmentation has led investigators to rely on time-consuming manual assessments prone to inter- and intra-rater variability. This study aims to develop an automatic T1L segmentation approach, adapted from a T2L segmentation algorithm. MATERIALS AND METHODS: T1, T2, and fluid-attenuated inversion recovery (FLAIR) sequences were acquired from 40 MS subjects at 3 Tesla (3 T). T2L and T1L were manually segmented. A Method for Inter-Modal Segmentation Analysis (MIMoSA) was then employed. RESULTS: Using cross-validation, MIMoSA proved to be robust for segmenting both T2L and T1L. For T2L, a Sørensen-Dice coefficient (DSC) of 0.66 and partial AUC (pAUC) up to 1% false positive rate of 0.70 were achieved. For T1L, 0.53 DSC and 0.64 pAUC were achieved. Manual and MIMoSA segmented volumes were correlated and resulted in 0.88 for T1L and 0.95 for T2L. The correlation between Expanded Disability Status Scale (EDSS) scores and manual versus automatic volumes were similar for T1L (0.32 manual vs. 0.34 MIMoSA), T2L (0.33 vs. 0.32), and the T1L/T2L ratio (0.33 vs 0.33). CONCLUSIONS: Though originally designed to segment T2L, MIMoSA performs well for segmenting T1 black holes in patients with MS.

MIMoSA: An Automated Method for Intermodal Segmentation Analysis of Multiple Sclerosis Brain Lesions.

BACKGROUND AND PURPOSE: Magnetic resonance imaging (MRI) is crucial for in vivo detection and characterization of white matter lesions (WMLs) in multiple sclerosis. While WMLs have been studied for over two decades using MRI, automated segmentation remains challenging. Although the majority of statistical techniques for the automated segmentation of WMLs are based on single imaging modalities, recent advances have used multimodal techniques for identifying WMLs. Complementary modalities emphasize different tissue properties, which help identify interrelated features of lesions. METHODS: Method for Inter-Modal Segmentation Analysis (MIMoSA), a fully automatic lesion segmentation algorithm that utilizes novel covariance features from intermodal coupling regression in addition to mean structure to model the probability lesion is contained in each voxel, is proposed. MIMoSA was validated by comparison with both expert manual and other automated segmentation methods in two datasets. The first included 98 subjects imaged at Johns Hopkins Hospital in which bootstrap cross-validation was used to compare the performance of MIMoSA against OASIS and LesionTOADS, two popular automatic segmentation approaches. For a secondary validation, a publicly available data from a segmentation challenge were used for performance benchmarking. RESULTS: In the Johns Hopkins study, MIMoSA yielded average Sørensen-Dice coefficient (DSC) of .57 and partial AUC of .68 calculated with false positive rates up to 1%. This was superior to performance using OASIS and LesionTOADS. The proposed method also performed competitively in the segmentation challenge dataset. CONCLUSION: MIMoSA resulted in statistically significant improvements in lesion segmentation performance compared with LesionTOADS and OASIS, and performed competitively in an additional validation study.